Affordable Access

Access to the full text

Intervention and Identifiability in Latent Variable Modelling

Authors
  • Romeijn, Jan-Willem1
  • Williamson, Jon2
  • 1 University of Groningen, Faculty of Philosophy, Oude Boteringestraat 52, Groningen, 9712 GL, The Netherlands , Groningen (Netherlands)
  • 2 University of Kent, Department of Philosophy Cornwallis North West, Canterbury, Kent, CT2 7NF, UK , Canterbury (United Kingdom)
Type
Published Article
Journal
Minds and Machines
Publisher
Springer Netherlands
Publication Date
Mar 30, 2018
Volume
28
Issue
2
Pages
243–264
Identifiers
DOI: 10.1007/s11023-018-9460-y
Source
Springer Nature
Keywords
License
Green

Abstract

We consider the use of interventions for resolving a problem of unidentified statistical models. The leading examples are from latent variable modelling, an influential statistical tool in the social sciences. We first explain the problem of statistical identifiability and contrast it with the identifiability of causal models. We then draw a parallel between the latent variable models and Bayesian networks with hidden nodes. This allows us to clarify the use of interventions for dealing with unidentified statistical models. We end by discussing the philosophical and methodological import of our result.

Report this publication

Statistics

Seen <100 times